DESCRIPTION: (Applicant's Description) Lung cancer will remain a major cause of mortality for a large part of the next century. The inverse association of this disease with fruit and vegetable intake is now well established, providing an opportunity for identifying the components (phytochemicals) in these foods which are protective. Many phytochemicals have shown cancer inhibiting activity in experimental models. Because of difficulty in recalling past dietary intake, and because of the lack of food composition data for a large number of these chemicals, human studies which use a prospective design and biomarkers of exposure are of great interest. Past dietary studies, including ours, have been helpful in suggesting which of these phytochemicals may be particularly important in humans. We propose to collect biospecimens on a large existing cohort and conduct a nested case- control study testing associations between lung cancer and markers of intake for several phytochemicals. A blood and overnight urine sample will be collected through Cores B and C from about 39,000 male and female Oahu residents of Caucasian, Japanese or Native Hawaiian origin who participate in the Multi-ethnic Cohort Study. Samples will be processed and stored in liquid nitrogen. Lung cancer cases will be identified through rapid reporting by our SEER registry. After an average of 3 years of follow-up, we expect 194 lung cancer cases. Two respiratory cancer-free controls will be matched to each case on age at phlebotomy, sex, race, month of sample collection and smoking status. Inverse associations of lung cancer with urinary total phenol and isothiocyanates, plasma carotenoids, tocopherols and ascorbic acid, and serum selenium will be tested, as well as a direct association with a marker of oxidative stress (urinary F2-isoprostane). These associations will be adjusted for urinary cotinine, a biomarker of smoking dose. The study should provide useful insight into the dietary etiology of lung cancer.